CIS-1355 · Security Engineering · Module 13

Incident Preparation and Investigation

Keep critical services running during disruption, restore them to defined targets, and preserve evidence of what happened.

Prepare for disruptions and test the plans

Business continuity plan (BCP)
Explains how essential business functions will continue during a disruption, such as by switching to manual intake or moving work to another location.
Business impact analysis (BIA)
Identifies essential functions, dependencies, single points of failure, and the impact of downtime or data loss. Use it to set recovery priorities and targets.
Disaster recovery plan (DRP)
Defines how to restore IT services and in what order. Restore dependencies such as identity, networking, and storage before the applications that rely on them.
Incident response plan
Assigns roles, escalation paths, communications, and procedures for detecting and analyzing incidents, containing damage, removing the cause, recovering, and applying lessons learned.

A tabletop is a discussion of a scenario. In a walkthrough, participants step through their procedures and handoffs. A simulation lets responders practice actions in a controlled scenario. A failover test checks that a service can switch to a standby system or site. In parallel processing, the recovery environment handles a test workload alongside production so the team can compare outputs without transferring production service. Record gaps, owners, and corrective actions after each exercise.

Use a framework such as the NIST Cybersecurity Framework to organize risk management and incident response. Use an adversary behavior knowledge base such as MITRE ATT&CK to map observed tactics and techniques and design realistic exercises.

NIST · Incident Response Recommendations · CISA · Tabletop Exercise Planner Handbook (PDF) · MITRE ATT&CK · Resources

Build resilience and set recovery targets

Server clusters can keep a service available when a member fails. RAID levels that use mirroring or parity can tolerate certain drive failures. RAID 0 stripes data for performance and provides no redundancy. Storage area network (SAN) multipathing keeps access available when a storage path fails, provided the paths do not share the failed component. Redundant network devices and links, an uninterruptible power supply (UPS), and a generator can cover different failure modes. A UPS bridges a short outage or gives equipment time to shut down cleanly. Check which dependencies the backup power actually supports.

A hot site is equipped for rapid activation. A warm site has some infrastructure and needs additional setup or recovery. A cold site primarily provides space and utilities and takes longer to activate. In the cloud, consider failure domains, backup and data locations, and whether multiple providers would reduce a shared dependency. Test failover and recovery against the service’s requirements.

Recovery point objective (RPO)
The maximum tolerable data loss measured in time. A 30-minute RPO calls for a usable recovery point no more than 30 minutes before the disruption.
Recovery time objective (RTO)
The target time to restore an acceptable level of service after a disruption. Measure actual recovery time in an exercise or incident and compare it with the target.

Backups are recoverable copies that can be made on a recurring schedule and stored offsite, offline, or in protected immutable storage. Replication maintains another copy, synchronously or with some lag, and can propagate deletion or corruption. A snapshot captures a point-in-time view and may depend on the original storage. A change journal or transaction log can support replay or point-in-time recovery if the technology and retention settings allow it. Test restores and measure recovery time and data loss. Keep an independent backup that survives a failure or compromise of the primary environment.

NIST · Contingency Planning Guide (PDF) · CISA and partners · StopRansomware Guide (PDF)

Reconstruct an incident from several sources

Preserve relevant records before they expire or are overwritten. Normalize time zones and account for clock offsets. Record gaps and distinguish observations from conclusions. Use the investigation to improve defenses and support any required reporting. Coordinate reporting decisions with the appropriate incident, legal, or compliance owners.

NIST · Incident Response and Cybersecurity Risk Management

Collect and document digital evidence

Digital forensics collects, examines, analyzes, and reports on digital evidence, including routine records and hidden, altered, or deleted data. E-discovery concerns identifying, preserving, collecting, reviewing, and producing relevant electronically stored information for a legal matter.

  1. Secure and document the scene. Confirm authority and scope. For a physical scene, record equipment, connections, and device state, and label relevant items and cables before moving them when feasible.
  2. Preserve and acquire evidence. Prioritize sources by volatility, value, and collection risk. Memory or active sessions may disappear on shutdown. Use suitable memory acquisition, disk imaging, or export tools, and protect offline media from writes where supported. Record what each collection captured and verify acquired data when possible.
  3. Examine. Work from a verified copy where practical. Use tested tools to locate artifacts such as authentication records, process traces, deleted files, or system settings.
  4. Analyze. Correlate findings, consider other explanations, and document the limits of the evidence.
  5. Report. Record acquisition methods, dates, tool versions, hashes where applicable, findings, and the evidence supporting each conclusion.

Provenance describes the evidence’s origin and collection context. Start the chain of custody at collection and record possession, transfers, storage, and authorized access as applicable. Keep the evidence and its documentation protected throughout the investigation.

NIST · Guide to Integrating Forensic Techniques into Incident Response (PDF)

Practice case

Ransomware disrupts customer intake. At 10:00, a ransomware incident takes the ticketing service offline. The team can keep receiving requests through an approved manual intake process. The ticketing service has a 30-minute RPO and a four-hour RTO. The latest candidate backup is from 09:15, and the standby replica contains affected data. Several endpoints are still running.

How would you keep intake working and contain the incident? If the 09:15 backup is the newest clean point, does it meet the RPO? By what time does the ticketing service need to be restored to meet the RTO? What evidence would you preserve?

Review a possible answer

Activate manual intake and track requests for later reconciliation. Isolate affected systems under the response plan and coordinate preservation of important volatile data with containment. Collect endpoint, identity, ticketing, and network records, and document their origin and custody.

The backup is 45 minutes before the disruption, so it misses the RPO by 15 minutes. Check for a later verified clean recovery point or trustworthy records that can reconstruct missing tickets. The RTO calls for service by 14:00. Restore to a clean environment, validate service and data before reopening it, and record any missed target.

Read and watch